Back

PoSTcode: Probabilistic image-based spatial transcriptomics decoder

Gataric, M.; Park, J. S.; Li, T.; Vaskivskyi, V.; Svedlund, J.; Strell, C.; Roberts, K.; Nilsson, M.; Yates, L. R.; Bayraktar, O.; Gerstung, M.

2021-10-14 bioinformatics
10.1101/2021.10.12.464086 bioRxiv
Show abstract

Realising the full potential of novel image-based spatial transcriptomic (IST) technologies requires robust and accurate algorithms for decoding the hundreds of thousand fluorescent signals each derived from single molecules of mRNA. In this paper, we introduce PoSTcode, a probabilistic method for transcript decoding from cyclic multi-channel images, whose effectiveness is demonstrated on multiple large-scale datasets generated using different versions of the in situ sequencing protocols. PoSTcode is based on a re-parametrised matrix-variate Gaussian mixture model designed to account for correlated noise across fluorescence channels and imaging cycles. PoSTcode is shown to recover up to 50% more confidently decoded molecules while simultaneously decreasing transcript mislabeling when compared to existing decoding techniques. In addition, we demonstrate its increased stability to various types of noise and tuning parameters, which makes this new approach reliable and easy to use in practice. Lastly, we show that PoSTcode produces fewer doublet signals compared to a pixel-based decoding algorithm.

Matching journals

The top 8 journals account for 50% of the predicted probability mass.

1
Briefings in Bioinformatics
◐ 354 papers in training set
Top 0.9%
7.7%
Similar papers in this journal
2
Genome Biology
○ 637 papers in training set
Top 1%
7.7%
Similar papers in this journal
3
Nature Communications
○ 5641 papers in training set
Top 21%
7.7%
Similar papers in this journal
4
PLOS Computational Biology
● 1863 papers in training set
Top 6%
6.6%
Similar papers in this journal
5
Bioinformatics
◐ 1204 papers in training set
Top 4%
6.6%
Similar papers in this journal
6
Nature Methods
○ 385 papers in training set
Top 2%
6.1%
7
Scientific Reports
○ 3612 papers in training set
Top 16%
5.4%
8
BMC Bioinformatics
○ 457 papers in training set
Top 2%
5.4%
50% of probability mass above
9
Bioinformatics Advances
◐ 203 papers in training set
Top 1%
4.8%
10
NAR Genomics and Bioinformatics
◐ 242 papers in training set
Top 1.0%
4.2%
11
GigaScience
◐ 212 papers in training set
Top 0.9%
4.0%
12
PLOS ONE
● 5266 papers in training set
Top 37%
3.2%
13
Biological Imaging
◐ 15 papers in training set
Top 0.1%
3.1%
14
Nucleic Acids Research
◐ 1281 papers in training set
Top 7%
2.3%
15
Cell Reports Methods
○ 165 papers in training set
Top 1%
1.9%
16
Methods
○ 34 papers in training set
Top 0.3%
1.7%
17
Communications Biology
○ 993 papers in training set
Top 17%
1.5%
18
Computational and Structural Biotechnology Journal
● 242 papers in training set
Top 4%
1.4%
19
iScience
○ 1154 papers in training set
Top 28%
1.1%
20
Forensic Science International: Genetics
○ 26 papers in training set
Top 0.1%
1.0%
21
Advanced Science
○ 286 papers in training set
Top 8%
1.0%
22
Cytometry Part A
○ 33 papers in training set
Top 0.3%
0.8%
23
Nature Machine Intelligence
○ 70 papers in training set
Top 2%
0.8%
24
Genome Research
● 468 papers in training set
Top 6%
0.8%
25
Biology Methods and Protocols
◐ 61 papers in training set
Top 2%
0.8%
26
IEEE Journal of Biomedical and Health Informatics
● 37 papers in training set
Top 1%
0.6%

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.